Development of Creative Commons licenses

[James Grimmelmann <james AT grimmelmann.net>]

Jake McKee wrote:
> Hey all,
> 
> Thanks so much for the great conversation. Rather than reply inline to a 
> lot of the points raised, I'll just summarize my response.
> 
> Generally I don't understand the fear of "branching". As someone 
> mentioned, the more options with something like CC, the more clarity. 
> Again, CC is about helping people "express intent". Offering more 
> options/attributes is a fantastic way to do that. (Of course there's 
> absolutely a point of diminishing returns, but two flavors of NC doesn't 
> begin to come close to that point)

Branching is one of my biggest fears with CC licenses, so I'll list a
few more reasons to be concerned about it.

(1) Every new license needs either to be supported for a long time or at
some point explicitly dropped.  The latter can create transition
problems -- people using a license may be stuck with older and older
versions, so that users have to read more and more carefully the
language of the actual legal deeds.  This happens, for example, with
anyone who refused to switch from SA 1.0 to BY-SA when the no-BY option
was phased out.  Thus, branching as an experiment has the consequence
that some licensors get stranded on branches that don't continue.  So
creating new licenses causes trouble if you ever want to go back to
fewer licenses.
(2) Branching complicates CC's communication mission.  Quick: which of
the following licenses are identical: Sampling Plus, Founders Copyright,
Attribution-ShareAlike, Attribution-NonCommercial, Wiki, and Music
Sharing?  Especially when the naming convention gets away from the
license elements, more licenses means a harder time figuring out what a
given license does and doesn't allow, and a harder time designing clear
communication schemes to explain the licenses.  It's easier to design a
 comprehensible icon set when the licenses are fewer and clearer.  (Just
look at the chaos of the Adobe CS3 icon set, and on the difficulty of
knowing what programs you get when you buy various versions of the CS3
suite.)
(3) Branching creates more and more incompatible pools.  With SA, those
pools are explicitly self-propagating and can't be combined in
perpetuity.  Even without it, the pools have a way of replicating
themselves.  If I want to use something under BY-NC, and to release my
contribution under a CC license, by far the easiest thing for me to do
is to choose a BY-NC license for my own additions.  That's a de facto
pool that doesn't combine nicely with, say, BY-SA.
(4) Branching creates more opportunities for strange interactions.  The
discussion here about mechanicals under various licenses is a good
example.  You have to figure out both what NC and SA would suggest about
whether to waive them -- and then to decide which takes precedence in an
NC-SA license.

There are, as you say, some good reasons to want more licenses.  My
first reaction is always to ask (a) whether there is an existing license
that approximates what someone wants closely enough, and (b) whether
there is a compelling case for changing an existing license.

Here, I think that the NC license provides an okay basis for a "no
resale" license in practice; use the NC and then state that you are also
okay with companies using it for internal purposes.  That gives the
general public an easy CC license for personal and all noncommercial
uses, but requires companies (which are generally in a better position
to research and negotiate licensing issues) to come back and check with
you before internal use.  That's not ideal, but it's not terrible.

On the other hand, I suspect that many people who use NC really mean
only something more like "no resale" and thus that changing the license
is not out of the question.  It'd make the license more useful on
average.  There are reasons not to (e.g. previous users relying on it
not allowing internal commercial uses, and many users who don't want to
allow them), but it still strikes me as being more useful than branching
a new license.

James